Mathias Silva Vazquez*, Tizie Bene**
Kenza Elass : kenza.elass[at]univ-amu.fr
Camille Hainnaux : camille.hainnaux[at]univ-amu.fr
Daniela Horta Saenz : daniela.horta-saenz[at]univ-amu.fr
Jade Ponsard : jade.ponsard[at]univ-amu.fr
*This presentation will summarize a new econometric approach for parametric income distribution analysis. Two recent strands of literature are bridged under this approach. Firstly, the literature on parametric corrections for income data sources evidenced to be prone to the problem of 'missing rich' such as household surveys. A new 'missing rich'-expanded parametric Generalized Lorenz curve form is proposed to account for a variety of high-income data issues such as under-reporting, top-coding, and/or undersampling. Secondly, a step-by-step procedure is introduced for the purpose of Bayesian inference on this model's parameters, including optimization of the correction quantities/margins for 'missing rich'. Procedures for the ellicitation of prior probabilities for the model's parameters, computational algorithms for parameter inference, and posterior checks for model criticism are proposed based on recent Approximate Bayesian Computation (ABC) methods for simulation-based Bayesian inference. The performance and computational properties of this approach to inference are illustrated through grouped- and micro-data examples on simulated data and on EU-SILC data.
**In many developing economies, formal insurance companies struggle to settle. People often mitigate risk by making insurance arrangements with each other, thus creating informal risk-sharing networks.Whereas in developed economies we observe the opposite situation. People rely more on formal insurance to cover their risks and less on informal links. A very recent empirical literature suggests that the introduction of formal insurance products affects informal risk-sharing among individuals. Either it strengthens the solidarity or it reduces it.
This work-in-progress is the first to tackle this issue from a theoretical viewpoint. To do so, we use a network-based approach. I introduce formal insurance into the model of formation of risk-sharing networks by Bramoullé and Kranton (2007). To the benchmark where individuals commit to sharing monetary holdings equally with a linked partner, I add the possibility for each agent to take out formal insurance. Preliminary results establish a relation between the structure of the risk-sharing networks that emerged and the situation in the insurance market.